Are bankruptcies the right metric to measure a CEO? For example, startup founders as a cohort probably lead a lot more companies to bankruptcy (or disbandment)

Are bankruptcies the right metric to measure a CEO? For example, startup founders as a cohort probably lead a lot more companies to bankruptcy (or disbandment) than laundromat owners. Yet probably the average startup owner is producing companies of higher value than laundromat owners (adjust for the right startup cohort to prove this point, like YC startups). Wouldn't you want to run a fair horserace like: - Randomly select MBA CEOs and non-MBA CEOs in a vintage year like 2000. - See the average earnings growth during their tenure, or stock price appreciation, or employee satisfaction, etc. He does cite two academic articles, but the first looks only at CEOs that have been featured on the cover of magazines, and in neither studies are the output variables the ultimate metrics that matter. (E.g. earnings management is fine if there are enough other positive qualities to redeem the CEO -- like getting long run growth for example).